Applicability of Paleoflood Surveys to the Black Hills of Western South Dakota

Transcription

1 Connecting South Dakota and the Nation South Dakota Department of Transportation Office of Research SD X Applicability of Paleoflood Surveys to the Black Hills of Western South Dakota Study SD Executive Summary Prepared by Jim E. O Connor and Daniel G. Driscoll U.S. Geological Survey 1608 Mt. View Road Rapid City, SD September 2007

2 DISCLAIMER The contents of this report reflect the views of the authors who are responsible for the facts and accuracy of the data presented herein. The contents do not necessarily reflect the official views or policies of the South Dakota Department of Transportation, the State Transportation Commission, or the Federal Highway Administration. This report does not constitute a standard, specification, or regulation. ACKNOWLEDGEMENTS The SD Technical Panel is composed of: Mark Clausen...FHWA John Costa... USGS Kevin Goeden... Bridge Design Kevin Griese...Materials & Surfacing Dennis Johnson...Research Dan Johnston...Research Richard Phillips... Bridge Design Steve Schelske...Rapid City Region The work was performed in cooperation between the South Dakota Department of Transportation (Office of Research) and the U.S. Geological Survey.

3 TECHNICAL REPORT STANDARD TITLE PAGE 1. Report No. SD X 2. Government Accession No. 3. Recipient's Catalog No. 4. Title and Subtitle Applicability of paleoflood surveys to the Black Hills of Western South Dakota 5. Report Date September Performing Organization Code 7. Author(s) Jim E. O Connor and Daniel G. Driscoll 9. Performing Organization Name and Address U.S. Geological Survey 1608 Mt. View Road Rapid City, SD Work Unit No. 11. Contract or Grant No. 12. Sponsoring Agency Name and Address South Dakota Department of Transportation Office of Research 700 East Broadway Avenue Pierre, SD Type of Report and Period Covered Draft Executive Summary September 2005 to August Sponsoring Agency Code 15. Supplementary Notes 16. Abstract Flood-frequency analyses for the Black Hills area have large uncertainties because of several complicating factors, including: effects of the massive 1972 storm near Rapid City; geologic influences; and potential influences of topography on precipitation patterns. The objective of this study was to assess the applicability of paleoflood hydrology techniques to generate better historical records on the magnitude and frequency of peak-flood events in the Black Hills area. As the study evolved, it became apparent that the paleohydrologic approach of greatest potential utility is detailed stratigraphic analysis of sequential deposits of slackwater flood sediments in caves, alcoves, and rock shelters, along with radiocarbon dating of entrained organic materials. This approach has been used worldwide in suitable environments to assess the frequency of large and infrequent floods and was applied at French Creek and Spring Creek, where detailed analyses demonstrated the regional applicability of this approach. Stratigraphic records for multiple sites in each stream reach provided excellent chronologies of previous large floods, indicating various flood events within the previous several millennia approaching or exceeding the approximate magnitude of 1972 flooding. Ensuing regional reconnaissance efforts indicated that local conditions are ideal for formation and preservation of flood slackwater deposits, especially within Paleozoic carbonate rocks ringing the periphery of the Black Hills, where abundant caves and alcoves protect deposits from erosion. Generally arid conditions aid in preservation of stratigraphic boundaries and detrital organic materials necessary for reconstructing flood chronologies. The overall conclusion of this reconnaissance-level study is that improved understanding of flood frequencies for the Black Hills region would result from implementation of future studies using established paleoflood techniques. Flood slackwater deposits have been identified in canyon sections of most major drainages in the Black Hills. These deposits show stratigraphic records of large floods that can be effectively used with radiocarbon dating and hydraulic analysis to determine the approximate timing and discharge of previous large floods. Specific flood issues that could be addressed by Black Hills paleoflood studies include (1) determination of the frequency, magnitude, and spatial characteristics of 1972-scale floods for several or all of the major Black Hills drainages; and (2) evaluation of spatial patterns of large-flood generation that owe to topography, geology, and climatology. 17. Keywords Paleoflood hydrology, slackwater flood deposits, floodfrequency analysis, Black Hills, South Dakota 19. Security Classification (of this report) Unclassified 20. Security Classification (of this page) Unclassified 18. Distribution Statement No restrictions. This document is available to the public from the sponsoring agency. 21. No. of Pages Price i

4 PROBLEM, PURPOSE, AND OVERVIEW OF TASKS Flood-frequency analyses for the Black Hills area have large uncertainties because of several complicating factors, including (1) effects of the extraordinary 1972 storm near Rapid City; (2) geologic influences; and (3) potential influences of topography on precipitation patterns. Methods relying on analysis of existing peak-flow data (numbers of sites and available periods of record) are insufficient to address these complications. Paleoflood hydrology techniques have been used successfully to improve peak-flow frequency relations in various areas; however, the applicability of such techniques is unknown for the Black Hills area. SDDOT Research Project SD was implemented as a cooperative reconnaissance-level study between the U.S. Geological Survey (USGS) and South Dakota Department of Transportation (SDDOT) to evaluate the utility of various potential paleoflood hydrology techniques for this area. The study was designed to address a single objective, specifically to assess the applicability of paleoflood survey techniques to generate better historical records on the magnitude and frequency of peak-flood events in the Black Hills area. This objective was designed to address two primary project purposes: (1) to evaluate the suitability of various paleoflood survey techniques for conditions within the Black Hills area; and (2) to provide a mechanism for scoping of potential future study phases, assuming that suitable paleoflood survey techniques are determined to be applicable for the area. Thirteen specific tasks were identified to guide research directions for this study. Tasks 1 and 7 involved initial and follow-up meetings with the Technical Panel; these tasks are addressed within the full completion report, but are not discussed within this Executive Summary. Tasks 2 through 5 were designed to locate and review existing information that might be relevant to initiation of a paleoflood study specifically for the Black Hills area. Task 6 was designed to identify candidate sites for trial application of appropriate paleoflood techniques. Tasks 8 through 11 were designed for testing of applicable techniques. Tasks 12 and 13, which involved communication of results through a final report and a presentation to the Research Review Board, also are not discussed within this Executive Summary. TASK 2 LITERATURE REVIEW A detailed literature search produced only two documents that addressed paleoflood hydrology for the Black Hills area, and neither document provided substantial insights for this study. Literature regarding various paleoflood hydrology techniques and application of such techniques in other areas is relatively abundant, and numerous relevant references are included within the full completion report. TASK 3 PERFORM INTERVIEWS Interviews were conducted with numerous individuals potentially having knowledge relevant to the study. The interviews did not yield much definitive information regarding the performance of paleoflood 1

5 surveys during this reconnaissance study phase; however, the following contacts could be useful for implementation of future potential study phases: The South Dakota State Archaeological Society (SAS) could be a very useful partner for future collaborative efforts, from several possible standpoints. Many of the sites with valuable paleoflood evidence are within small caves or are located in areas where caves are abundant; thus, future coordination with the Black Hills Grotto Club could be very useful. Numerous contacts were made with area residents and landowners, and additional communications would be a critical component of future potential study phases for obtaining: (1) access permission when working within private land holdings; and (2) various information regarding past flow events or other items of interest. It is noteworthy that public interest in this study has been very high, probably because of the relevance and interesting nature of the subject matter. Public visibility has large potential to provide many important benefits for potential future study phases. TASK 4 SCRUTINIZE 1972 PEAK-FLOOD DATA It was determined that the 1972 peak-flood estimates have been thoroughly examined by various USGS hydraulic experts as part of the standard review process and as part of various subsequent efforts. Cursory additional examination of all available measurement files indicated that sound procedures were consistently used in processing and reviewing the original data, and that confidence is high that most estimates are reasonably reliable. Data available for stream reaches where detailed future investigations might occur would be useful for various purposes, and closer examination of applicable files would be appropriate as part of potential future study phases. TASK 5 INVESTIGATE FLOOD RECORDS FOR STATIONS UNAFFILIATED WITH USGS Much information is available regarding historical flooding in various locations; however, historical accounts tend to be sketchy and much of the information would be useful only for documenting dates and general magnitudes of flood events. More exhaustive investigations would be useful for detailed paleoflood surveys for specific stream reaches and additional efforts in this area would be appropriate for potential future study phases. TASK 6 SELECT CANDIDATE SITES FOR PALEOFLOOD SURVEYS This task initially focused on identifying locations where paleoflood evidence existed that could be amenable for testing various possible techniques of paleohydrologic analysis. As the study evolved, it became apparent that the paleohydrologic approach of greatest potential utility was detailed stratigraphic 2

6 analysis of sequential deposits of slackwater flood sediments in caves, alcoves, and rock shelters, along with radiocarbon dating of entrained organic materials. This approach has been used worldwide in suitable environments to assess the frequency of large and infrequent floods and was applied at French Creek and Spring Creek, where detailed analysis demonstrated the regional applicability of this approach (as described in tasks 7 10). Thus, later efforts of site selection focused on evaluating the potential for application of this highly successful approach on a regional scale and resulted in a list of visited stream reaches throughout the Black Hills area where fully implemented application would address the key issues of flood-frequency and spatial and topographic controls on flood generation noted in Section 1.0 in the full completion report. Flood slackwater deposits typically form in zones of flow separation and are composed of sand and silt deposited from the suspended load of individual floods. Within caves and alcoves, or where otherwise protected from erosion or extensive disturbance by plants and animals, successive floods may leave a sequence of deposits forming a stratigraphic record of large flow events encompassing many hundreds or thousands of years. Individual flood deposits typically are separated by rockfall, organic duff, or local tributary sediment (which commonly has distinctly different grain size and mineral composition than other deposits), commonly allowing for determination of the number of floods that left deposits at a particular site. Organic detritus, including charcoal, driftwood, and blown-in leaf fall within and between individual flood deposits, can be dated using radiocarbon techniques, allowing for determination of the approximate timing of individual floods as well as the length of record represented by a sequence of flood deposits. The elevations of individual flood deposits indicate a minimum value for the elevation or stage attained by the flow. By combining stratigraphic and chronologic records of multiple depositional sites with hydraulic flow modeling techniques for obtaining corresponding discharge estimates, robust flood-frequency analyses are possible that vastly improve recurrence interval estimates for rare, high-magnitude floods. A key finding is that local conditions are ideal for formation and preservation of flood slackwater deposits, especially within Paleozoic carbonate rocks ringing the periphery of the Black Hills, where abundant caves and alcoves protect deposits from erosion. Upstream metamorphic and intrusive rocks provide abundant sand carried by larger floods, with mineral compositions distinct from those of slopewash and fluvial sediment derived from local drainages and hillslopes. The general arid conditions within rock shelters aid in preservation of stratigraphic boundaries and detrital organic materials necessary for reconstructing flood chronologies. These ideal conditions provide an unparalleled opportunity for reconstructing flood histories. For evaluating the applicability of various stream reaches as candidates for future potential paleoflood surveys, two categories of drainage basins were identified: (1) relatively large drainage basins having stream reaches entrenched in canyons within outcrops of the Paleozoic carbonate rocks around the periphery of the Black Hills sites in these basins have potential for providing improved frequency estimates of 1972-scale floods; and (2) drainage basins in the vicinity of the Limestone Plateau area within the higher elevations of the Black Hills, which would be useful for addressing the role of topography and geology in producing large runoff events. Reconnaissance efforts resulted in development of a list of stream reaches in both categories where paleoflood evidence has been identified. 3

7 A primary finding from these extensive reconnaissance efforts (in combination with other results of this study) is that flood-frequency issues for the Black Hills area could be substantially resolved through full implementation of comprehensive paleoflood studies in strategic locations. The suitability of any particular slackwater deposit for a comprehensive paleoflood survey (e.g. number of flood layers and availability of organic material for radiocarbon dating) cannot be determined until it has been excavated to full depth; however, the detailed analyses in the two sampled reaches (described in tasks 8 10) indicate that appropriate alcoves commonly contain deposits of several large floods that occurred during the last several hundred to several thousand years. Many of the canyon areas with potential for alcoves are large, and more extensive searches would be needed in combination with preliminary excavations to locate the most ideal combinations of alcoves for comprehensive paleoflood surveys. However, these reconnaissance efforts have clearly demonstrated that alcoves with good potential for conducting successful paleoflood surveys are available within most if not all of the largest drainages around the periphery of the Black Hills area. Reconnaissance was conducted for 15 streams within the aforementioned category of large drainage basins, and one or more alcoves potentially containing stratigraphic records of large floods were located in each of these drainages. All of these streams have relatively large drainage areas, which is advantageous because paleoflood evidence typically is more abundant in larger drainages where larger flood flows typically are generated. Also, a relatively small number of large drainages compose a substantial part of the entire Black Hills area; thus, high-resolution data could be obtained for much of the area with surveys in relatively few basins. In addition, systematic peak-flow records are available for one or more applicable gaging stations within most of these drainage basins, which would allow incorporation of paleoflood results with existing peak-flow data. Various evidence from 1972 flooding (including slackwater deposits and deposits of coarse sediments, flotsam, and other debris) is available in about one-half of the larger drainages. Such evidence provides an excellent point of reference for both stratigraphic and hydraulic analyses and would be useful in conducting future paleoflood studies in affected stream reaches. Substantial reconnaissance also was conducted for the second general category of drainage basins in high-elevation areas near or within the Limestone Plateau, where paleoflood evidence is less common because (1) drainage areas generally are relatively small; and (2) geologic conditions generally are much less favorable for the existence of suitable alcoves than in locations around the periphery of the Black Hills. Several drainages were located (fig in the full completion report) where sufficient paleoflood evidence was found (including some alcoves with flood sediments) to conclude that a future paleoflood study to address the questions regarding the higher-elevation areas could be successfully implemented. TASK 8 SAMPLE THE MOST PROMISING SITES On the basis of reconnaissance investigations and guidance from the technical panel, more detailed analyses were conducted on two reaches: (1) a ~1-mile-long segment of Spring Creek, upstream from U.S. Highway 16 where it flows across the Minnelusa Formation; and (2) a ~1-mile-long segment of French Creek within Custer State Park. These sites were selected because they each contained numerous alcoves and caves floored by slackwater deposits and met criteria of being variously affected by the 1972 flow that 4

8 were established through discussions with the Technical Panel. The analyses at each of these two sites, as elaborated in the following section (tasks 9 and 10), included the following: Detailed searches along canyon walls for slackwater deposits in areas protected from erosion or other disturbance. Preliminary shallow excavations to tentatively evaluate the clarity of stratigraphic records and number of preserved events. Selection of multiple alcoves at different elevations for detailed stratigraphic analysis. Excavation and detailed stratigraphic description and recording of slackwater deposits and preserved flood flotsam. Samples of organic materials for radiocarbon dating. Site surveys to document elevations of deposits and cross-section geometry for preliminary discharge estimations. TASKS 9 AND 10 CONDUCT COMPREHENSIVE PALEOFLOOD SURVEYS AND AGE DATING Comprehensive paleoflood surveys were conducted along Spring Creek, for which substantial flooding occurred in 1972, and French Creek, for which a large peak has not occurred within the period of systematic record. Both reaches are within streamflow loss zones where streamflow seldom occurs, which has been a factor in minimizing human disturbance of flood deposits in sheltered environments. These surveys consisted primarily of (1) excavating, describing, and sampling flood sediments at selected sites; (2) age dating selected excavated materials using radiocarbon ( 14 C) analyses; and (3) performing cursory surveys of channel cross sections for calculation of preliminary discharge estimates. Detailed surveys along Spring Creek were conducted in four primary locations, informally named the Hailstorm, Superscour, Olympia, and Temple of Doom Alcoves. Results for two pits at the Hailstorm Alcove tentatively show that three individual floods recorded by the stratigraphy span a time period extending back to about 400 to 200 B.C. Radiocarbon dates from the penultimate flood deposits (immediately below the 1972 deposits) at two pits indicate that this flood, which was probably larger than the 1972 flood on the basis of deposit thickness and its coarse texture, was sometime between A.D and The Hailstorm Alcove site apparently is so high that it only rarely is inundated, thus recording only the largest floods of the last several thousand years on Spring Creek. Four 14 C analyses from three pits at Superscour Alcove indicate that the record here is shorter than at Hailstorm Alcove, but contains evidence of more floods of smaller magnitude during the last several centuries. The basal 14 C date for Pit 1A indicates that the sequence of four thin Spring Creek flood deposits plus the overlying and thicker 1972 deposit accumulated since A.D Here, the penultimate flood deposits in both Pits 1A and 1B were apparently emplaced sometime after A.D A single radiocarbon analysis from near the base of Olympia Alcove gives a similar result as Pit 1A 5

9 at Superscour Alcove the record at Olympia Alcove tentatively shows that six floods (including 1972) left deposits in this relatively low alcove since A.D A relatively thick and coarse deposit at the base of the Olympia Alcove deposit may be equivalent to the similar-aged large-flood deposit preserved below the 1972 deposit at Pit 3A at Hailstorm Alcove. An analysis of a twig within a driftwood accumulation above the highest 1972 flotsam near the Temple of Doom Alcove gave a calibrated age of A.D This driftwood is higher than any flotsam from the 1972 flood. Additional flood sediments also were found about 5 feet (ft) higher at this site and indicate a discharge of at least 26,000 cubic feet per second (ft 3 /s), relative to a 1972 discharge of ~22,000 ft 3 /s in this reach. Detailed surveys along French Creek were conducted in two locations, informally named Pratt Cave and the Riviera site, which is an exposed tributary fan. Evidence within Pratt Cave was exceptional and included three sequences of slackwater deposits for three large floods (exceeding ~9,000 ft 3 /s) and driftwood at several locations many feet higher than the flood sediments. The highest driftwood at Pratt Cave was a 10-ft-long log lodged about 33 ft above the channel thalweg that gave a date of A.D This corresponds to a discharge of ~92,000 ft 3 /s for a drainage area of ~105 square miles (mi 2 ). Comparable 1972 examples for the unregulated portions of drainage areas along Rapid Creek include 31,200 ft 3 /s for station (above Canyon Lake, unregulated area of 51 mi 2 ) and 50,000 ft 3 /s for station (at Rapid City, unregulated area of 90 mi 2 ). The basal date from a lower pit outside the entrance to Pratt Cave indicates floods achieved stages of at least 3.6 to 6 ft above the French Creek thalweg since A.D TASK 11 FROM ALL LOCATIONS SAMPLED, DETERMINE APPLICABILITY OF PALEOFLOOD DATA FOR IMPROVING FLOOD-FREQUENCY ESTIMATES The analysis of historical flood records (gaged and otherwise), in combination with regional reconnaissance of most large Black Hills drainages and subsequent detailed studies at the French Creek and Spring Creek sites, indicates that application of paleohydrologic techniques has excellent potential for substantial improvement of flood-frequency estimates for many streams in the Black Hills region. This finding is especially true for particularly large, low-frequency floods such as those that occurred in several drainages in This finding owes to several considerations, data sets, and information sources obtained during this study: 1. The historical and gaged record of flooding shows that exceptional floods can result from large thunderstorm complexes affecting substantial portions of the Black Hills. The largest measured floods are associated with 1972 flooding, with flood peaks on the order of as much as about 50,000 ft 3 /s for several different streams with contributing drainage areas in the range of about 50 to 150 mi Such floods on individual drainages are rare, relative to systematic record lengths, with at most one or two major floods having affected the larger Black Hills drainages since the late 1800s (as determined from tasks 2 5). Some drainages have no record of major floods during this 6

10 timeframe. The few historical occurrences of these events make it exceedingly difficult to reliably judge their frequency from gaged and historical observations alone. Additionally, their relative rarity inhibits reliable inferences of large-flood generation processes that may vary spatially (due to factors such as topography, geology, or climatology) over the Black Hills region. 3. Preliminary analysis of flood slackwater deposits at Spring Creek and French Creek (reported in tasks 8, 9, and 10) show that these stratigraphic records of flooding are resolvable into confident interpretations of the number, timing, and magnitude of large floods, including flows perhaps larger than the maximum flows of the 1972 flood. Moreover, these stratigraphic records extend sufficiently far back in time (commonly more than 2,000 years) and contain records of enough floods so that robust flood-frequency estimates, particularly for large flows such as those of 1972, can be determined from established statistical techniques. 4. The availability of flood evidence from 1972 flooding, including slackwater deposits, deposits of coarse sediments, flotsam, and other debris, would be highly useful in conducting future paleoflood studies in affected stream reaches. Such evidence provides an excellent point of reference for both stratigraphic and hydraulic analyses. 5. The regional reconnaissance conducted in this study under task 6 shows that alcoves, rock shelters, and caves within canyon sections of most major drainages and some smaller streams preserve stratigraphic records of large floods. The broad distribution of streams with such stratigraphic records could allow for addressing questions of spatial control of 1972-scale floods. 6. Paleoflood evidence is less abundant in the smaller drainage areas that dominate the higherelevation parts of the Black Hills area. Sufficient paleoflood evidence, including flood slackwater deposits, does exist in strategic locations, however, such that questions regarding geologic and topographic controls in the higher-elevation areas probably could be successfully addressed. FINDINGS AND CONCLUSIONS The overall conclusion of this reconnaissance-level study is that improved understanding of floodfrequency for the Black Hills region would result from implementation of future studies using established paleoflood techniques. Flood slackwater deposits are preserved in caves, alcoves, and rock shelters in canyon sections of most major drainages in the Black Hills. These deposits record stratigraphic records of large floods that can be effectively used with radiocarbon dating and hydraulic analysis to determine the approximate timing and discharge of previous large floods. Specific flood issues that could be addressed by Black Hills paleoflood studies include (1) determination of the frequency, magnitude, and spatial characteristics of 1972-scale floods for several or all of the major Black Hills drainages; and (2) evaluation of spatial patterns of large-flood generation that owe to topography, geology, and climatology. Addressing these issues would substantially improve understanding of processes for flood generation and the frequency of large floods in the Black Hills region. The following are specific findings and conclusions: 7

11 1. Rare and large floods such as the 1972 flooding leave distinctive suites of deposits, such as flotsam accumulations and slackwater deposits. 2. In protected environments such as caves, rock shelters, and alcoves, such deposits accumulate for several centuries or millennia, forming stratigraphic records of large floods that can substantially extend historical and gaged records of flooding. 3. At French Creek, for which the historical record of floods is sparse with no known flows of 1972 magnitude, slackwater deposits and driftwood accumulations at Pratt Cave recorded relatively large flows over the last ~700 years, including at least two major floods with discharges exceeding ~35,000 ft 3 /s and one flow 700 to 1,000 years ago with a discharge of ~90,000 ft 3 /s. 4. At Spring Creek, where the 1972 flood attained a peak discharge of ~22,000 ft 3 /s, there is evidence of at least three flows of 1972 magnitude during the last 2,400 years, including one larger flow years ago. 5. At both Spring Creek and French Creek, slackwater deposits at different elevations contain stratigraphic records spanning different time periods, with low accumulations recording smaller floods over the last several hundred years, and higher deposits accumulating deposits from only the largest floods over the last 2,000 4,000 years. 6. These types of results from the preliminary analyses at Spring and French Creeks, if additionally supported by more geochronology and precise hydraulic modeling, can be used in conjunction with established statistical analysis procedures to substantially improve flood-frequency estimates for floods of the general rarity and magnitude of the 1972 flooding. 7. Our regional reconnaissance (task 6) indicates that the types of flood deposits found along Spring and French Creeks are preserved in most, if not all, major drainages, indicating a high probability that robust flood-frequency estimates for large and infrequent floods such as the 1972 flood could be determined for many of the large drainages in the Black Hills region. 8. Additionally, our reconnaissance has located similar slackwater flood deposits in caves, alcoves, and rock shelters flanking many smaller and higher-elevation Black Hills drainages, indicating that paleohydrologic approaches, if applied widely across the region, have high potential to address questions of geologic and topographic controls on the frequency large and rare floods in the Black Hills. APPLICABILITY OF PALEOFLOOD METHODS FOR FUTURE STUDIES The reviews, analyses, and conclusions arising from the task definitions for this research study resulted in the following summary of applicability of paleoflood methods for future studies and anticipated outcomes regarding potential implementation of future study phases: 8

12 (1) Full implementation of future studies using established paleoflood analysis techniques would reduce uncertainty in frequency estimates for large-magnitude and infrequent floods for many applicable stream reaches in the Black Hills region. Preliminary comprehensive paleoflood investigations for Spring Creek and French Creek demonstrated the quality of highresolution flood chronologies that can be obtained in appropriate environments. Extensive reconnaissance efforts have shown that similar potential exists within many other applicable stream reaches. (2) The best approach for reconstructing high-resolution flood chronologies is by detailed stratigraphic analysis of flood slackwater deposits found in the abundant caves, alcoves, and rock shelters in canyon settings entrenched in Paleozoic carbonate rocks around the periphery of the Black Hills. These sheltered areas preserve stratified flood deposits and entrained organic materials for hundreds and, locally, thousands of years. Stratigraphic interpretation and dating of these flood-deposit sequences can provide long-term chronologies of large flood events. Additionally, the wide range of alcove elevations provides stratigraphic records of large floods spanning different magnitude and age ranges, thus providing relatively comprehensive records of floods over a wide range of magnitudes. Various other forms of paleoflood evidence also can contribute useful information in many locations. (3) Future studies phases, if implemented, would warrant inclusion of detailed hydraulic and statistical analyses. Detailed surveys of channel geometry throughout extensive stream reaches would be required to support rigorous hydraulic analyses needed to make adequate estimates of previous large flood discharges. Utilization of appropriate statistical analytical procedures incorporating infrequent large flood events would be needed for development of robust floodfrequency estimates. (4) Studies focused on the larger drainages where they flow through the carbonate rocks on the periphery of the Black Hills, such as the Spring Creek and French Creek sites studied in detail for this phase, would provide the most information on the size and frequency of scale flows. Most if not all of these larger drainages in the Black Hills have sites with high potential for paleoflood analysis. Regional reconnaissance indicates that slackwater flood deposits are available in most of the larger drainages, and evidence from 1972 flooding would provide a valuable point of reference in many of these drainages. Results for this setting would be most meaningful for assessing hazards for the majority of populated areas and infrastructure concentrated along the flanks of the Black Hills. (5) Studies of higher-elevation and smaller drainages would provide important information regarding flood generation processes and controlling factors such as elevation, topography, and geology. Smaller drainages in the higher elevations generally have few flood deposits preserved, but reconnaissance in the Black Hills indicates several possible sites for which detailed investigations probably would provide substantial information on flood-frequency and magnitude. Such information would contribute to site-specific flood-frequency analyses as well as broader frequency analyses within the Black Hills region that use regionalization techniques. 9